• State Resolved
  • Locked Locked
  • Replies 17 replies
  • Subscribers 48 subscribers
  • Views 3912 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Problem on XTR117 directly driven by current source

mike Zhang
Prodigy 100 points
Other Parts Discussed in Thread: XTR117, OPA333

Hi,  I am making 4-20mA circuit.

      output stage is xtr117 driven by 40uA zero + 0-160uA(current output from OP)..

     I found that it takes too much time(10 minutes or more) to back to stable 4mA output .

    40uA id coming from 3V and 75K(These are very accurate).

   Any explaination on this.

Thanks

Mike

 

  • 0
    TI__Guru* 98065 points
    Mike,


    I'm moving this to the Precision Amplifiers Forum. I'm sure it could go into the /etc Amplifiers and other Linear Forum, but it should get some attention here as well.


    Joseph Wu
  • 0
    TI__Guru 64055 points

    Hello Mike,

    Could you elaborate on your statement:  "I found it takes too much time (10 minutes or more) to back to stable 4mA output."?

    Is the output current from the XTR117 oscillating?  If so, at what frequency, and could you share an image of the output waveform?  Or, do you mean that the dc output current takes a long time to settle when the input changes?

    Also, please share a full schematic so we can understand and verify the circuit connections.

  • 0 in reply to Collin Wells
    Prodigy 100 points
    Thanks for your attention and help.


    XTR117 is not oscillationg, it happens only one time while power on or at the warm-up stage.

    and starting value of the output is not very certain. some tome 4.5mA, some time 4.4mA......when I power off at 4.21mA and immediately power on again, it continues to warm up that is it starts to drop from 4.21mA again.......


    I did lower the capacitor value at the output of PIn8(+5V output) from 10uF to 1uF, it did help to improve the time of warm up.


    the circuit schematic will be added here late today or tomorrow when i come back.


    any way I can state here:


    Xtr117 +5V output connected to 1uF capacitor, driving two voltage reference 3V(capacitor 10uF) and 1.2V(capacitor 10uF)

    Iin= 40uA coming from 3V(voltage reference +75K).

    += 0uA ------ 160uA coming from the precision amplifier with current output.





    Best regards


    Mike
  • 0 in reply to mike Zhang
    TI__Guru 64055 points
    Hi Mike,

    Thanks for the additional information and for sending the schematic later today. After we analyze the circuit connections we'll make some suggestions to improve the circuit performance.
  • 0 in reply to mike Zhang
    Prodigy 100 points
    How can attach the pdf file or image here?

    Mike
  • 0 in reply to mike Zhang
    TI__Guru 64055 points

    Hi Mike,

    First, click on the "Use Rich Formatting" button at the bottom right of the reply area to use the rich formatting options for the E2E post.  

    Then use click on either the "Insert Image" or "Insert File" buttons shown in the image below:

  • 0 in reply to mike Zhang
    TI__Guru 64055 points

    Hi Mike,

    Thanks for the additional information. 

    What connects to the P1 4-pin connector?  It looks like a resistive bridge or other sensor that is powered from 1.2V.  How much current does this sensing element consume from the 1.2V supply?  If it's a resistive bridge sensor, what is the bridge resistance?

    We've seen issues in the past from connecting large capacitors between the V+ and IRET pins of the XTR11x devices such as C6 in your schematic. Was this the capacitor you reduced from 10uF to 1uF and saw improved results?  

    The issues we've seen in the past were related to decreased AC output current bandwidths which prevented the proper transmission of some higher frequency sensor signals which is much different than the extremely slow start-up behavior as you're experiencing. 

    Have you considered powering your LDO circuits from the +5V VREG output of the XTR117?  There aren't any issues associated with placing the 1 - 10uF input capacitors for the two LDOs between VREG and IRET.  Doing so would also reduce the power dissipation of those LDOs by reducing the voltage drop across them.

  • 0 in reply to Collin Wells
    Prodigy 100 points
    P1 4-Pin connector is 350Ohm bridge power by 1.2V as we can see.
    C6 has been reduced from 10uF to 1uF.
    max9934 supply voltage is restricted and I can not power it with +5V REG
    and Bridge sensor supply use 1.2V much accurate than +5V REG.
    I did not reduce the value of C61, C62 as I measured the 3.0V and 1.2V both stable while output loop current is droping.
    ....
    I could not find the reason why loop current drops to stable 4.00mA while power on in such long time.
  • 0 in reply to mike Zhang
    TI__Guru 64055 points
    Hi Mike,

    Thanks for the additional information.
    My previous suggestion was to power the 1.2V and 3V LDOs from the +5V REG output. This means that the MAX device would still be powered from the +3V LDO output. I suggested this modification because as I mentioned above placing a capacitor from V+ to IRET interacts with the frequency performance of the XTR11x devices.

    That said, I think the issue may be related to the power dissipation of the circuitry powered from the loop. I recently wrote a blog about this topic which you can read here:

    e2e.ti.com/.../two-wire-4-20ma-transmitters-background-and-common-issues-part-3

    Basically, if the circuitry powered from the loop dissipates more than ~3.5mA then the transmitter will not be able to regulate the output down to the proper zero-scale level as described in the blog. The fact that your issue seems to present itself at the 4mA zero-scale level agrees with this theory.

    My main concern is that the 350 Ohm bridge will require 3.42mA of the limited <3.5mA current budget. This may not be leaving enough current to properly power the 1.2V and 3V LDOs along with the MAX9934 current sensor. I'm not sure why the issue seems to fix itself over time but try placing 1k resistors in series with 350Ohm bridge as suggested in the blog to see if reducing the bridge current positively impacts the power-on time of this circuit. I understand that you'll likely have to change the gain of your sensing circuitry to make up for the decreased output span from the sensor due to the addition of the 1k resistors.
  • 0 in reply to Collin Wells
    Prodigy 100 points
    P1 4pin connector connects to 350 OHM bridge powered by 1.2V
    C6 capacitor has been reduced from 10uF to 1uF already.
    C61,C62 has not yet reduced the value, as I checked the output voltage 1.2V and 3.0V OK while power up.
    +5V REG can not used for voltage restriction on max9934 and sensor bridge accuracy.
    ...........
    Still i do not know the reason why loop current takes more than 10 minutes to drop to normal 4.00mA

    mike
  • 0 in reply to Collin Wells
    Prodigy 100 points

    Thank you for your information

       Now I figure out the possible reason:   Power dissipation of XTR117 REG is limited at 3.5MA as you said.

    But I read the datasheet Xtr117 REG should have the ability of 4mA as shown in this chart.

    So my design is based on the above chart and set  Maximum current =3.95mA for all circuit at all temperature.

    By the way,the suggested circuit( INA OP, or OP) in your blog seems not working to me, because Vos is too much to me.

    I will redesign the circuit and follow the REG limit current<=3.5mA.

    Best regards

    Mike

  • 0 in reply to mike Zhang
    TI__Guru 64055 points

    Hi Mike,

    You're correct, the VREG output can source more than 4mA.  However, it is up to the designer of the 2-wire transmitter to limit the quiescent current of their sensor conditioning circuitry to less than the desired zero-scale output (3.5mA, 3.8mA, 4mA, 5mA, etc.) or the transmitter will not be able to regulate the output down to the zero-scale level as described in the blog.

    The 3.5mA level that I quoted in the previous blog was a conservative number.  You should base the maximum current your conditioning circuitry can draw on the datasheet quiescent current specifications of the XTR117 which I've copied below.  Over the full operational temperature range of the XTR117, the maximum quiescent current is 250uA. Therefore the conditioning circuitry can draw up to 3.75mA of current and the output will be able to regulate down to 4mA. 

    However we usually recommend some margin in case the initial power-up current of the circuitry is higher than expected which would impact start-up behavior of the transmitter. 

    I looked at the Vos of the amplifiers listed in the blog and the OPA333 features the same maximum offset voltage as the MAX9934.  May be worth taking a look at.

  • 0 in reply to Collin Wells
    Prodigy 100 points
    Hi, Mr. Collins Wells.

    Great thanks!
    This information is so valuable to me as I have to consider the accuracy as well as temperature.
    And now this XTR117 start-up problem is clearly understood.
    Based on the limit of 3.75mA, I will have the new circuit soon.

    Best regards
    Mike
  • 0 in reply to mike Zhang
    TI__Guru 64055 points
    Great! Let us know if you have any additional questions.